The overall objective of the proposed research is to test the efficacy of computer delivered personalized normative feedback in the prevention and reduction of problem drinking among college students. Despite the wide spread utilization of social norms prevention approaches, methodological confounding has prevented determination of the magnitude and duration of effect of personalized normative feedback. In addition, this research aims to evaluate gender specificity of the normative referent, individual differences in self-determination, and reasons for drinking as moderators of intervention efficacy. This research will also determine the impact of normative information on students who never or rarely drink. College student drinking is largely motivated by social factors, including the perception that heavy drinking is "normal." College students overestimate the alcohol consumption of their peers. This distortion is presumed to cause many students to drink more than they otherwise would. It may also contribute to feelings of alienation and isolation among students who never or rarely drink. Correction of normative misperceptions is the goal of social norms prevention approaches. This research will be conducted at two universities in 3 stages. In stage 1, descriptive drinking norms from both sites will be obtained and incorporated into a web based assessment and intervention program. Stage 2 will implement a 3-year longitudinal efficacy trial among heavy drinking students. Students will be randomly assigned to a single exposure of personalized normative feedback, repeated biannual exposure to feedback, or a control group. Students in the feedback groups will also be randomly assigned to receive gender specific or gender non-specific feedback. Stage 3 will evaluate the impact of normative information on never/rarely drinking students. Students will be randomly assigned to receive social norms marketing messages, personalized normative feedback, or assessment only. Follow-up assessments will determine the impact of normative information at 3-months and 6-months post baseline. All assessments and interventions in stages 2 and 3 will be Internet based. This research will fill important knowledge gaps regarding the use of normative feedback in preventing and reducing problematic drinking and will serve as a prototype for implementing low-cost internet based personalized normative feedback interventions. [unreadable] [unreadable]